Quantile Advantage Entropy Safe Reasoning

Stabilize LLM reasoning training by replacing mean-based advantage baselines with K-quantile baselines, preventing both entropy collapse and explosion while improving performance on mathematical benchmarks through response-level gating and asymmetric sample weighting.

adu2021 Updated 7 repo stars

File contents

adu2021/skillxiv commit 08b093087a

Frequently asked questions

npx get-skill add adu2021/quantile-advantage-entropy-safe-reasoning